Factors affecting consumers' intention towards ...

Journal of Food Products Marketing

ISSN: 1045-4446 (Print) 1540-4102 (Online) Journal homepage: http://www.tandfonline.com/loi/wfpm20

Factors affecting consumers’ intention towards purchasing halal food in South Africa: a structural equation modelling Abdalla M. Bashir, Abdullah Bayat, Samuel Oladipo Olutuase & Zul Ariff Abdul Latiff To cite this article: Abdalla M. Bashir, Abdullah Bayat, Samuel Oladipo Olutuase & Zul Ariff Abdul Latiff (2018): Factors affecting consumers’ intention towards purchasing halal food in South Africa: a structural equation modelling, Journal of Food Products Marketing, DOI: 10.1080/10454446.2018.1452813 To link to this article: https://doi.org/10.1080/10454446.2018.1452813

Published online: 22 Mar 2018.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=wfpm20

JOURNAL OF FOOD PRODUCTS MARKETING https://doi.org/10.1080/10454446.2018.1452813

Factors affecting consumers’ intention towards purchasing halal food in South Africa: a structural equation modelling Abdalla M. Bashira, Abdullah Bayata, Samuel Oladipo Olutuasea, and Zul Ariff Abdul Latiffb a

School of Business and Finance, University of the Western Cape, Cape Town, South Africa; bFaculty of Agro Based Industry, University of Malaysia Kelantan (UMK), Jeli, Kelantan, Malaysia ABSTRACT

KEYWORDS

This research study highlighted the factors that influence the intention of consumers to buy halal food products. In this study, the researchers have chosen to address the gap in the literature pertaining to non-South African consumers’ purchase intention. Non-South Africans are foreign people who reside in South Africa including immigrants, workers, and students. Through utilising the theory of planned behaviour, this study aims to determine the attitudes, subjective norms, perceived behaviour control, and awareness of non-South African towards purchasing halal food products. This also attempts to identify the strongest factor that influences the purchasing intention of halal food by non-South African consumers. The sample of the study consisted of 230 nonSouth African consumers in Cape Town. Data were collected through self-administered questionnaires. Structural equation modelling (SEM) was used to identify the structural relationships among the constructs identified and to test the study hypotheses. The findings of this study indicate that there is a significant relationship between awareness and buying behaviour. Interestingly it found that only the attitude dimension of TPB had a significant relationship with purchase intention, whereas subjective norms and perceived behavioural control did not show a significant relationship with the consumer intention variable.

Halal awareness; Intention; Buying behaviour; Halal food products; TPB; South Africa

Introduction Food industries have become more concerned and sensitive about the manufacturing and sourcing of halal products. Halal products broadly receive acknowledgement as a scale for food safety and quality assurance (Majid, Abidin, Majid, & Chik, 2015). Currently, the concern for consuming halal food products is not only prevailing among Muslim consumers, but also the halal awareness has extended to non-Muslims as well. Awareness of halal food is growing worldwide particularly in non-Muslim countries including South Africa. For this, there has been a greater awareness among hotels, guesthouses, CONTACT Abdalla M. Bashir [email protected] University of the Western Cape, Faculty of Economic and Management Sciences, School of Business and Finance, Private Bag X 17, Bellville 7535, Cape Town, South Africa. Tel: 27 21- 959 3332, Fax: 27 021 959 3219 Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/wfpm. © 2018 Taylor & Francis

2

A. M. BASHIR ET AL.

and corporate venues and they have included halal menus. Additionally, the success of the halal food industry depends on the awareness of halal consumer (Mohamed, Shamsudin, & Rezai, 2013). Trade in halal products is estimated at around US$150 billion annually (Bamba, Talek, & Kaba, 2017; Egan, 2002) and growing. According to the Holy Quran, a halal food product is a product that meets the requirements of halal according to Islamic law. Most foods and drinks are considered halal except that it has been obviously stated as non-halal (haram) or prohibited in the Holy Quran and the prophetic hadith. Therefore, Allah (SWT) has commanded His Messengers Muhammed (S.A.W), where He says: “O mankind! Eat of that which is lawful and good on the earth. . .” (Al-Quran. Surah Al-Baqarah, Chapter II Verse 168). Halal food and beverages are described as anything that can be eaten or drunk and there is no lawful evidence about its prohibiting, as well as their ingredients are free from any prohibited or contaminated components (Said, Hassan, Musa, & Rahman, 2014). Halal products are food being produced and consumed not only in countries that have high population of Muslims, but also in countries that have a low population of Muslims including South Africa, this makes halal products available worldwide (Hassan, 2013). Trade of halal foods market is growing as one of the markets with tremendous potential in the world (Berry, 2008; Muhammad, Isa, & Kifli, 2009; Shah Alam & Mohamed Sayuti, 2011; Sungkar, Othman, & Hussin, 2008), which makes a trade in halal products globally successful (Hassan, 2013). Research conducted by Halal Industry Development Corporation (2014) and Halal Research Council (2010) estimated that the global halal products market is valued at USD2.3 trillion (not including Islamic finance and banking), which means that halal industry is a lucrative trillion dollar business. Of this trade, the halal food and beverages represent USD1.4 trillion, with 67% share of this market (The Halal Journal, 2010). In addition, the demand for halal products and services has increased to US $655 billion in 2010 from US$635 billion in 2009 (Bernama, 2009). Halal food products are a concern for consumers mainly Muslims, because of its hygienic, cleanliness, animal welfare, and safety issues (Marzuki, Hall, & Ballantine, 2012), and because it provides further assurances on food safety and quality (Nastasha, 2015; Teng, Siong, & Mesbahi, 2013). This is so because there is a high demand for halal products in some non-Muslim countries for both Muslim and non-Muslim consumers (Berry, 2008). This is the case in South Africa, in which Muslims are a minority. Although Muslims being a minority in South Africa, it has a highly developed halal ecosystem. In South Africa, halal products are broadly available in local market (SalaamGateway.com, 2017). It also represents about 50% of the total consumption of the productive sector in the African continent (Bamba et al., 2017). However, in South Africa, there are no formal statistics on halal food production or sales, but South African National Halaal Authority (SANHA) estimated that in 2012, 60% of all food products were halal-certified (SalaamGateway.com, 2017). In addition, the Western Cape provincial government is researching and

JOURNAL OF FOOD PRODUCTS MARKETING

3

seriously considering the establishment of a halal agroprocessing food park in Cape Town (Hancock, 2015). Halal products are also growing in popularity due to the real awareness that halal products are healthier, safer and is produced using humane animal treatment (Teng et al., 2013). For these reasons, there is a need for researchers to investigate halal consumer intention and behaviour of Muslim and non-Muslim in South Africa. However, in this research study, we focus on resident non-South African Muslim and non-Muslim consumers’ intention and behaviour in the Western Cape.In South Africa, the awareness of halal amongst Muslims and non-Muslims has not yet investigated by South African scholars. In addition, the intention of consumers towards buying halal food has also not been investigated thus far, there is also another category of halal consumers involved in consuming halal food products which has not yet been investigated, which is the category of non-South African consumers whether Muslims or non-Muslims. This category constitutes about 72,184 of the population group of Cape Town city with the percentage of 1.93% (Census, 2011). Since there are no reliable statistics with relation to these consumers intention towards buying halal food products, we decided to focus on this category in our research.

Review of literature and hypotheses Halal food products in South Africa

South Africa is one of the five largest manufacturers of halal products worldwide and broadly regarded as a gateway to the continent’s halal food and beverage market. South Africa’s halal industry has changed it into a Sub-Saharan Africa’s halal powerhouse, with its halal business valued at $3.22 billion (SalaamGateway. com, 2017). Although the population of Muslims in South Africa is only around 2.5% of the country’s total population, there is an increased focus in meeting the needs of these Muslim consumers. Furthermore, South Africa contributes 25% of the African continents’ total gross domestic product (GDP). Despite its small percentage of Muslims, which is 1.2 million, South Africa has emerged as a leader in halal food industry because of the presence of highly advanced halal certification programmes. Although South Africa before 1994 was governed under the Apartheid Regime, the respect of religious practice was not restricted. The demand for halal products and services is increased amongst the Muslim population, mainly Muslims of Indian decent and Muslims known as Cape Malays, who consist of the decedents of Indonesian slaves brought to the Cape during colonial times is in this country. There is a growing population of Muslims converts, as well as Muslims from other countries such as Somalia and Bangladesh. Concerning the non-Muslim community, there is awareness in halal products. Primarily, it is believed to be the same as Kosher, and many of the nonMuslim South Africans are not sufficiently enlightened about the halal

4

A. M. BASHIR ET AL.

requirements. In addition, consumers particularly Muslims are very cautious in purchasing halal products particularly in relation to the food consumption, personal care products, hotels, and restaurants because of religion obligation. Although the value system of Muslim consumers differ from other non-Muslim consumers in relation to consume food products (Varinli, Erdem, & Avcılar, 2016). Muslim consumers are also similar to other consumers, demanding healthy and quality products, which necessarily comply with Shariah requirements (Al-Harran & Low, 2008). Moreover, Muslim consumers are more aware of the importance of consuming halal foods, which indirectly leads to the extension of global halal food industry (Man & Sazili, 2010). Theory of planned behaviour (TPB)

TPB is a theory suggested by Icek Ajzen that shows a relationship between beliefs and behaviour (Ajzen, 1991). TPB has been demonstrated to be the best way in predicting individual on consumption intention (Ajzen, 1991; Taylor & Todd, 1997), and has been applied to study the relations among attitudes, subjective norms, and perceived behaviour control to determine the intentions and behaviours of individuals (Ajzen, 2005). In addition, TPB provides a social-psychological framework to understand and predict the determinants of human behaviour and integrates some of the fundamental concepts in the social and behaviour sciences (Armitage & Conner, 2000). TPB was applied to examine how attitude, subjective norms, and perceived behaviour control can influence the intention. Subsequently, it influences the behaviour of non-South African consumers to buy halal food products. In addition to the TPB factors, the awareness of non-South African halal consumers, which represents another factor may affect the intention and afterwards the behaviour of non-South African consumers will be examined. TPB is considered as relevant in studying consumer preferences in buying food products as many researchers in previous studies (Abdul Latiff & Ayob, 2014; Rezai, Puah, Mohamed, & Shamsudin, 2012; Shah Alam & Mohamed Sayuti, 2011) had successfully applied it. It is a convenient model, which is useful in predicting the consumer purchasing behaviour. Many authors have recommended the efficiency of this model to measure halal purchase intention (Afendi, Azizan, & Darami, 2014; Khalek & Ismail, 2015; Zulariff & Mohamad, 2014). According to TPB, attitude towards the behaviour, subjective norms, and perceived behavioural control typically predict behavioural intentions. However, the research model proposed in this study shown in Figure 1 is based on TPB. The intention of halal food purchasing preceded the process before actual purchase or buying behaviour, as the intention reflects future behaviour. Accordingly, awareness has a relationship with buying intention and buying behaviour. Attitude, subjective norms, and perceived behavioural control all have direct relationships with buying intention. Finally, buying


5

Figure 1. Purchasing-intention-decision model.

intention has a relationship with consumers’ buying behaviour. Based on the proposed research model, there are five hypotheses to be tested. Figure 1 shows the research model and hypotheses tested in this study. Awareness of halal

Awareness of halal literally means having a particular interest in or experience of halal, besides being knowledgeable of halal foods, drinks, and products (Ambali & Bakar, 2014). Awareness refers to the understanding and information about specific things (Aziz & Chok, 2013). Additionally, it is a relative concept where a person possibly will be, to some extent, subconsciously or very aware of the issues relating to halal aspect of what is allowable by Allah (God). Kertajaya and Ridwansyah (2014) state that without prior knowledge of the product, there is relatively high probability that consumer will not intend to buy the product, whether these products are relevant to halal products or non-halal products. Moreover, the consumer’s awareness of halal affect their decisions towards buying halal food products (Zakaria, Abdul Majid, Ahmad, Jusoh, and Zakaria (2017). Aziz and Chok (2013) emphasised that halal awareness has an influence on clarifying the intention to purchase halal products. Furthermore, Hamdan, Issa, Abu, and Jusoff (2013) in their study reveal that most influential factor in Muslim consumers’ choices to buy halal food is the level of knowledge about dealing with the product. Although many studies have been conducted on the halal food industry and on halal consumer in Muslim majority countries (e.g. Afendi et al., 2014; Shah Alam & Mohamed Sayuti, 2011), only a few studies attempt to investigate the problems in measuring the factors that affect the intention of consumers in majority of non-Muslim countries. There has been no detailed published research studies on non-South African halal consumers’ intention as far as we are aware. Accordingly, the important research question in the study was as follows: How do individual factors influence halal consumers’

6

A. M. BASHIR ET AL.

intention and behaviour towards purchasing halal food products? To support the above research question, the next research sub questions are posed: Q1. Do awareness, attitude, subjective norms, and perceived behavioural control significantly affect the purchase intention of halal food products? Q2. Does awareness directly or indirectly affect the buying behaviour of halal food products? Q3. Does purchasing intention affect the buying behaviour of halal food products significantly?

However, the objectives of this study are: ●

● ●

To determine if awareness, attitude, subjective norm,s and perceived behavioural control significantly influence purchase intention towards halal products. To determine the direct and indirect effect of awareness on buying behaviour towards halal food products. To determine the significant effect of purchasing intention on buying behaviour of halal food products.

Awareness

Generally, awareness refers to the human perception and intellectual response to the condition of what they eat, drink, and use (Ambali & Bakar, 2014). Awareness is the procedure that is being taken in order to buy a product or service. It usually reflects the first step of the purchasing process, in which consumers who initially are not well versed with the product or service become familiar with it. In other words, awareness means the knowledge or understanding of particular subject or situation (Randolph, 2003). Previous research (Abd Rahman, Asrarhaghighi, & Abd Rahman, 2015; Aziz & Chok, 2013; Hamdan et al., 2013) have asserted that awareness has a positive influence on consumer attitude towards buying the food products. In addition, it has a positive effect on consumer intention (Krishnan, Aderis, Azman, & Kamaluddin, 2017; Zakaria et al., 2017). Accordingly, the study hypothesises that: H1a: Awareness about halal products has a positive and significant impact on consumer buying behaviour towards buying halal food products. H1b: There is a positive correlation between awareness and consumer purchase intention towards buying halal food products.


7

Attitude

Ajzen (1991) defines attitude towards a behaviour as “the degree to which a person has a favourable or an unfavourable evaluation of a behaviour in question”, it shows the performance of individual’s behavioural intentions are positively or negatively valued (Ajzen, 2005). Attitude was measured based on previous studies done by Aditami & Soepatini (2016) and Taylor & Todd (1995). According to previous studies, attitude has a significant and positive influence on purchasing intention, which is drawn from consumers’ positive attitude towards halal food products (Abd Rahman et al., 2015; Mukhtar & Butt, 2012; Shah Alam & Mohamed Sayuti, 2011). Therefore, more intention has been paid to purchase halal food products. Accordingly, the study hypothesises that: H2: There is a positive correlation between attitude and consumer purchase intention towards buying halal food products. Subjective norm

Subjective norm refers to the consumers’ perception towards social normative impetus (Ajzen, 2005), which may include family, friends, colleagues, relatives, or other such significant groups (Ajzen, 1991, 2005). These samples may offer the pressures on individuals. Subjective norm is the social pressure that a person feels to indulge or not indulge in a particular behaviour (Ajzen, 2005). Subjective norm is measured based on previous studies (Aditami & Soepatini, 2016; Armitage & Conner, 1999; Dean, Raats, & Shepherd, 2012). Prior research done by Bonne, Vermeir, Bergeaud-Blackler, & Verbeke (2007), Sukato (2008), and Yoh, Damhorst, Sapp, & Laczniak (2003) argued that consumers’ decision-making on product choice is certainly influenced by friends and family members. Accordingly, the study hypothesises that: H3: There is a positive correlation between subjective norm and consumer purchase intention towards buying halal food products. Perceived behavioural control (PBC)

Perceived behavioural control (PBC) refers to “people’s perception of the ease or difficulty of performing the behaviour of interest” (Ajzen, 1991). It is an individual’s ability to perform a given behaviour (Ajzen, 2005). At the same time, PBC influences the intention and behaviour both directly and indirectly through behavioural intention (Noar & Zimmerman, 2005). PBC comprises of three variables namely: ability, controllability, and availability. PBC was measured based on the previous studies (Aditami & Soepatini, 2016; Chen & Peng, 2012; Dean et al., 2012; Sparks, Guthrie, & Shepherd, 1997; Taylor & Todd, 1995). In addition to that, Ajzen (2005) emphasised that consumers need to

8

A. M. BASHIR ET AL.

uphold a certain degree of actual control in order to give themselves selfconfidence in carrying out the behavioural intentions. Due to the combination of intentions and perceived behavioural controls one can predict human behaviour (Ajzen, 1991). This leads to perceived behavioural control (Ajzen, 2002). Accordingly, the study hypothesises that: H4: There is a positive correlation between perceived behavioural control and consumer purchase intention towards buying halal food products. Consumer’s purchase intention and buying behaviour

Purchase intention refers to individual’s readiness and willingness to purchase a certain product or service (Ajzen, 1985), and it can influence the purchasing decision of consumers in the future (Omar, Mat, Imhemed, & Ali, 2012). This can be considered as one of the mechanisms of consumer cognitive behaviour on how a consumer intends to purchase a certain product (Hosein, 2012). Additionally, purchase intention is the direct originator of behavior (Ajzen, 1991). According to Ajzen (1985), purchase intention is the state where an individual is willing to perform a behaviour. Accordingly, Ajzen and Fishbein (1980) defined actual purchase behavior as the individual’s willingness to purchase specific manufactured goods or services. Purchase intention was measured based on the previous studies (Kumar & Mokhtar, 2016; Taylor & Todd, 1995). Several empirical studies also show that intention can be a reliable predictor of behaviour. Davies and Wright (1994) explain that buying behaviour comes from the adequate awareness towards product. Therefore, the study hypothesises that: H5: There is a positive correlation between purchase intention and buying behaviour of non-South African towards buying halal food products. Methodology Data collection and sample

A random sampling method was used. Randomised samples of 300 respondents were targeted, which represent different people and different geographical areas in Cape Town city. This sample size was suitable as it is suggested and the recommended value set by (Comrey & Lee, 1992; Hair, Black, Babin, Anderson, & Tatham, 2010). The questionnaire for this study was established based on the document that was presented on Ajzen’s website, which entitled “Constructing a TPB questionnaire” (Ajzen, 2002; Francis et al., 2004). This document helped in constructing the survey instrument relevant to this study. According to that, a selfadministrated questionnaire was utilised to collect information regarding


9

non-South African halal consumers’ purchasing intention and awareness towards buying halal food products. The questionnaires were distributed in different areas and places in Cape Town city, including restaurants, shops, Mosques, and public places such as gardens, libraries, and train station. The purpose of study and questions were explained to the respondents. Therefore, they could just fill the questionnaire. In addition, before handing in the questionnaires, the respondents were asked if they consume halal food products. After collecting the 230 questionnaires, these questionnaires were implicit and entered into Excel sheet for further analysis. A set of items to create questionnaires were administrated based on previous literature. The questionnaire was initially written in English and then translated into Arabic. Totally, 280 responses were received, out of which only 230 were found usable and comprehensive. Therefore, the response rate was 77%, and it shows appropriate statistical reliability and generalisability (Stevens, 2012). However, some 22 questionnaires were rejected because it had more than 10% missing values in dependent variables, which were not acceptable according to the statistics guidance (Hair, 2010), In addition, there were 28 discarded questionnaires which were filled by South African halal consumers while those were not targeted in this study. Whereas the rest of 20 questionnaires did not return. For data analysis, reliability, descriptive, structural equation modelling (SEM) analysis have been used. Analysis method

SPSS software version 24 was utilised in this study to analyse all the data. Reliability and descriptive analysis both were used to analyse the consistency and demographic information respectively. In addition, the hypotheses are tested by applying SEM, which is a relatively robust method of analysis. SEM represents a philosophy that differs significantly from that typically followed in marketing modelling (Steenkamp & Baumgartner, 2000). However, the questionnaire was divided into three sections. The first section comprised of the respondents’ socio-demographic characteristics, which includes age, gender, marital status, religion, nationality, education level, and occupation. The second section consisted of statements concerning non-South African halal consumers’ awareness. The third section was comprised of halal consumers’ attitude, subjective norms, and perceived behavioural control towards buying halal food products. Five Likert-type items which measured the response of the questionnaire was used which are Strongly Disagree (1), Disagree (2), Neutral (3), Agree (4), and Strongly Agree (5) (Vagias, 2006). Likert scale is a measurement scale used to measure attitudes, opinions, and one’s perception of social phenomena.

10

A. M. BASHIR ET AL.

According to Gwinner (2006), collective agreement from the forum stated that five (5) point scale is the most appropriate when conducting a survey. Research instrument

The instrument of this study was developed based on the TPB framework, which consists of three constructs namely attitude, subjective norms, and perceived behavioural control in addition to awareness. The items were developed using the guidelines of constructing questionnaire proposed by its originator Izek Ajzen. All items went through the process of modification and rephrasing in order to fit the context of non-South Africans’ halal consumption intention and behaviour. To ensure the content and construct validity of the research, two scholars seek agreement between a theoretical concept and a particular measuring procedure validated the questionnaire. In addition, the questionnaire was pre-tested by 20 respondents. In a quantitative research, validity refers to the degree to which a study precisely reflects or measures the specific concept that researcher is trying to measure. Therefore, to increase the validity and reliability of the results of this study, all the items have been adapted from previously validated studies (Straub, 1989) and any changes required to fit the instruments to the current sample context were appropriately performed. Based on rigorous instrument search, the reflective type measurement items for the study’s variables were adopted. Findings Reliability analysis

Cronbach’s alpha reliability coefficient was used to measure the internal reliability of the questionnaire, and the values obtained for each variable are illustrated in Table 1, which shows that all variables had reliability values of more than 0.8 (ranging from 0.873 to 0.945), thereby suggesting that the questionnaire items are reliable on their ability to measure the respective constructs consistently. Cronbach’s alpha according to Nunnally (1978) has indicated 0.70 to be an acceptable reliability coefficient. Table 1. Reliability Analysis of Constructs Reliability Analysis of Constructs Variable 1 2 3 4 5 6

Attitude Subjective norm Perceived behavioural control Awareness Purchase intention Buying behaviour

ATT SN PBC AW PI BB

No. of Items 5 4 4 3 3 3

Cronbach’s Alfa 0.932 0.904 0.945 0.874 0.873 0.877


11

Descriptive analysis Socio-demographic information In this study, descriptive analysis was used to describe the socio-demographic profile of the respondents. The study focused on Cape Town city in the province of Western Cape in South Africa, where 230 non-South African halal consumers that consists of (78.3%) males and (21.7%) females were involved. Among the respondents 56.5% of them are married and 43.5% are single. They are also grouped into six age group categories, which are below 21 years, (21–30) years, (31–40) years, (41–50) years, (51–60) years, and 60 years and above. By looking at Table 2, most of the respondents were at the age below 21 years old (35.2%). Although they are all non-South African consumers, but they have different beliefs as they have a different religion. Islam remains as the highest group amongst other religions (75.2%). This is because most of the participants in this study were Muslims. However, the second largest religious group is Christian (17%). All the respondents also came from various educational backgrounds. Out of 230 respondents, 77 (33%) were postgraduate students, followed by secondary students (27%). The lowest education level was primary students (11.3%). Regarding occupation, most of the respondents are students (37.8%), followed by selfemployed (27.8%). Finally, the nationality of the respondents, most of the respondents are Libyans (18.7%), followed by Egyptians (15.7%).

Hypotheses results

After the study is translated into the hypotheses model as shown in Figure 1, observed variables are drawn with the error terms for each latent variable. For the exogenous variables, attitude (ATT) contains five observed variables, while subjective norm (SN) and perceived behavioural control (PBC) each one contains four observed variables. Structural equation modelling (SEM)

The method of analysis employed in this study is the SEM, which was done via Amos and SPSS version 24. The SEM is used to test the effects in the relationships stated in research assumptions/questions (Cheung & Rensvold, 2002). Obtaining a fitted SEM is pertinent to achieving the objectives of this study. According to Hooper, Coughlan, and Mullen (2008) the following indices are important in realising a fitted structural model. Relative ChiSquare should be less than 3.00 (if sample size is less than 200); incremental fit index (IFI) should be less than 0.90; comparative fit index (CFI) should be less than 0.90; root mean square error of approximation (RMSEA) should be less than 0.08 and a PCLOSE greater than 0.05. Hooper et al. (2008) noted

12

A. M. BASHIR ET AL.

Table 2. Socio-Demographic Information of Respondents (N = 230) Demographic Variables Gender Age

Marital status

Religion

Occupation

Education Level

Nationality

Item Male Female Below 21yrs 21–30yrs 31–40yrs 41–50yrs 51–60yrs 61yrs and above Single Married Other Muslim Christian Buddhist Jewish Hindu Other Student Employee Unemployed Self-Employed Other Primary Secondary Undergraduate Post graduate Other Somali Libyan Sudanese Egyptian Moroccan Palestine Indian Bangladesh Saudi Arabian European countries Angolan Brazilian Yamani Nigerian Congo Namibian Eritrean Turkey Tanzanian Zimbabwean Burundian Gambian Malaysian Malawi Rwandan

Frequency 180 50 81 71 45 22 11 0 100 130 0 173 39 1 11 5 0 87 33 45 64 1 26 62 58 77 0 19 43 17 36 15 24 9 3 5 17 8 5 1 3 5 2 2 2 3 3 3 1 1 1 1

Percent (%) 78.3 21.7 35.2 30.9 19.6 9.6 4.8 0 43.5 56.5 0 75.2 17 4 4.8 2.2 0 37.8 14.3 19.6 27.8 0.4 11.3 27 25.2 33.5 0 8.3 18.7 7.4 15.7 6.5 10.4 3.9 1.3 2.2 7.4 3.5 2.2 0.4 1.3 2.2 0.9 0.9 0.9 1.3 1.3 1.3 0.4 0.4 0.4 0.4


13

that chi square index is sensitive to sample size, small or large. They therefore advised that caution should be exercised in the interpretation of the index. However, it is common practice to report the chi square index, but just ignore it for more reliable indices, when it comes to reporting on the goodness of fit for SEM. The structural model

Depicts the structural equation model for the conceptual model. From this model, there are six (6) variables involved in the study. Some factors as stated in Table 1 measure each of the variables. The boxes labelled as aw1, aw2, aw3, etc. represent the factors used to measure each construct (i.e. AW, ATT, SN, PBC, PI & BB); while the circles labelled as e1–e30 are the error terms associated with the factors. The single-headed arrows measure the regression weights while the double-headed arrows measure correlation/covariances. Model fitness and fitted model

Through a process of model fitness, some factors and variables were eliminated in order to realise a model that fits the dataset. The results which are presented in Table 3 show that the model has a good fit with the dataset and thus, its results are reliable. Fitted model results

Based on the fit indices discussed above, a fitted structural model was derived and the results of standardised regression weights and correlation estimates are presented in Table 4 and Table 5. From Table 4, we examined the statistical significance of each factors used to estimate the study variables such as awareness (AW), attitude (ATT), subjective norm (SN), perceived behavioural control (PBC), purchase intention (PI), and buying behaviour (BB). The results show that all the factors in the fitted model whose regression weights range between 0.797 and 0.936 are all statistically significant at p value less than 0.001. The regression weights also show high loading of Table 3. Model Fit Results Index Chi-Square (P value) Relative Chi-Square Comparative Fit Index (CFI) Incremental Fit Index (IFI) Root Mean Square Error of Approximation (RMSEA) PCLOSE

Obtained Value 0.000 2.283 0.953 0.954 0.075 0.00

Critical Value >0.05 3.00 0.90 0.90 0.05

Model Fit *Ignore Good Good Good Good Fair

*Chi-square fit index is usually only reported, but ignored when determining the fitness of a structural model because it is reactive to sample size.

14

A. M. BASHIR ET AL.

Table 4. Standardised Estimates of Regression Weights in the Fitted Structural Model PI PI PI PI BB BB a1 a2 a3 a4 a5 s1 s3 s4 s2 p1 p3 p4 p2 aw1 aw3 aw2 i1 i3 i2 b3 b1